[truthsurge]

I just bought some KRK Rokit 5 monitors to replace Cambridge SOundworks I had for 10 years (just low end PC speakers). got the KRKs in, installed and they have (by comparison) more mids, less highs and a bit more lows. I turned the high knob on back up to +1 db and the vol knob on back down by say 6db as they were just too loud and hummed noticeably. I think my old sound card is making some of that 60 cycle hum.

So, now, I have no idea if these speakers are giving me an accurate picture. I was happy w/old ones sans lack of bass/low end but I figured I should get a "real" set since I am doing more voice over and audio now.

SO, should I try to measure the speaker response and if it's too far off, get some other kind or should I measure the ROOM where I sit and then try to compensate for that with EQ (I don't even think my current sound card has any good EQ built in) and how in the can I do those two tasks??? I opened the software and it is not user friendly. It asked for SPL calibration and LUCKILY I had a decibel meter and I entered 85 db but that depended on my volume so I have no idea what I'm doing and why it seems like rocket science.

So, it's the old I have no idea if the old speakers were WEAK on midrange and I had the tone up for extra highs or if these new ones are just strong in the mids and soft in the highs.

Not to start a new paragraph with SO again, but... I am in a small say 15 x 15 bedroom with padding on walls and ceiling and carpet. It's pretty dead. I mainly was wanting some good flat monitors for my small room that would give me an accurate story of the sound going into them. In fact, these Rokit 5 are too loud for me. 45 w each and it is just more than I really want. They are putting out a lot more heat than those others which were like... 5 watts each? hahha But they sounded nice to me. I wonder if there are lower wattage ones that are still really flat and full in the lows. ?

Anyway, is there a GOOD tut on how to use REW to measure the actual speaker freq curve using an omni dir mic like AT-10 by Audio Technica?

Any tips or help is appreciated and don't hesitate to break it down to the basement level for me. I'm not dumb. I just need clear instructions.

Thanks!

TruthSurge

 

[robbo266317]

To get the speakers response you can close mic them and it will show what they are capable of.
Then you measure them "In Room" to see what areas you need to address to improve their overall response.

 

[truthsurge]

Thanks. So, is there a good tutorial on just how I can do that with that REW software? I understand the concept but just not the implementation of that software. It seems really unnecessarily difficult. Seems you should simply put mic say 1 ft away from speaker, turn OFF other speaker to avoid phasing or whatever then click a button, it sends that sine wave from 0 to 20KHz and records. then pops up the freq curve.

But just how to do that I have no idea with that software.

Anyway, I saw one tut by some company that just kept flashing it's logo infuriatingly throughout and I don't think I really saw a start to finish how to in that.

thanks!

ts

 

[Barleywater]

Read online help file!

and click button:

View attachment Measure Button.bmp

 

[EarlK]

:sn:

 

[EarlK]

- A related thread on ( the wish for ) tutorials .

:sn:

 

[truthsurge]

Thanks for the tut links. yeah, that one GIK or whatever was the one I already watched. Maybe I'll just need to watch it again and try to follow along but again, my sound card may not have these ins and outs so I'm not sure I can calibrate or test it but maybe I can skip that part. no idea.

anyway, I will read and watch again.

Thanks!


T

 

[truthsurge]

After tons of failures, I hired a team of monkeys who were able to figure out how to loop an output to an input and calibrate my soundcard and then do a measurement of the room and two more up close to my KRK rokit 5s and then the old speaker . But the graphs surprised me. There is a huge scooping of the mids or at least it SEEMS that way for maybe it's not the mids scooped but the lows and highs peaking. ?

Now, there is a natural 130 to 135 hump or resonant frequency in my room that I knew about just from using my ears. I routinely dipped that when processing my voice but maybe there are some other ones but also I wonder if the AT-10 mic I used is coloring it some. That mic has a fairly flat response and is omnidirectional.

Now, when I measured the room I put the mic facing the ceiling but when I measured each speaker i put it facing the speaker mid way between tweeter and woofer. But strangely, both graphs of the speakers look almost identical! How? They don't sound identical at all to me. The KRKs sound boxy with less highs or clarity so shouldn't that graph have a higher midrange??

Anyway, I did everything by the tutorial except I didn't have that same mic so I used the AT-10.

Now, I do see the hump at 135 in the graph so that makes sense to my brain. But why are the graphs almost the same if these speakers sound so diff???

Does anyone ever look at these if I could post them somehow?

Thanks again for all the help. I'm now wondering if I should just order a pair of A5X and compare side by side in THIS room and do another measurement and then compare with my ears to see which I like but I always thought the goal was to acheive a basically flat curve that perhaps rolled off slightly as the freq got higher. My graphs are NOT doing that and there is a really noticable hump from like 1K to 7K just going up up up from like 70db to 85db! The graph looks like a bowl with the mids scooped out.

I don't think my sound card has any EQ so I can't compensate even if I knew it was just the room but I had the mic right up to the speaker so I'm a bit confused at this point. I would be glad to post those 3 graphs but I know it's tiresome helping newbies all the time. I'm just beginning to think about this because I am trying to get some small speakers that work but aren't too large like those LSR 28P which were awesome but just too large for my room. Those (to me) sounded right. They sounded transparent and clear with no weird boxiness or anything. but gosh, they were too large for what I need.

Thanks!

TS

 

[truthsurge]

Question... seems the most noticeable problem my room has is the big hump around 130 to 135 Hz. Would a bass trap help eliminate that or is that too high for a bass trap? I think it's not but what do I know?

I did find a used pair of A5X Adam monitors that I think I will buy and then I should have enough data to figure out what is going on and which of the two monitors I like best but for some reason my gut says I'm going to like the Adam best due to it's flatter response and better tweeter. I am kind of a sucker for some clear high end. But I would like what I hear to be what is really there so others hear it too so I guess that's where eliminating the res frequencies of my room come in.

So, if anyone can throw more tips at me or look at those graphs, that'd be awesome.

thanks
ts

 

[robbo266317]

Now that you have 5 posts you should be able to use the "Go Advanced" button at the bottom of the text box and upload them and insert them.

 

[truthsurge]

Okay, I can upload them but let me say what I did since then.

I took another reading using my CAD M179 (a good large diaphram mic that lets me set it to omni pickup pattern) so I put it where my head sits in omni mode with no 100 hz rolloff. I measured and got a graph that looked similar but better somehow than the AT-10 mic. Didn't look as mid-scooped. Okay, so I then made an EQ offset for that to try to level it as close to flat as possible. Okay, so then I saved that to a TXT file so I could manually enter those parameters into a couple of EQ plugs that I could just set up directly in Reaper. I needed say 17 bands or so but I didn't worry with the stuff below 50 Hz.

Now, I entered it all perfectly. Played the tune Constant Craving by K.D. Lang. WITHOUT the EQ. it sounded good. I kicked in both plugs (i had 7 bands on one and the remaining 10 on the next) and it then sounded VERY hollow as if it was somehow all out of phase. like when you get the polarity wrong on wiring up your speakers.

So, I then just turned off the bands other than the huge dip at 100 and peak at 137 and it sounded... a little better but not better than no eq at all. Also same thing with the higher bands it just took all the life out of it instead of making it sound better.

So, my question is... why?

So, I made it so far only to be totally bummed out.

At this point, it makes no sense. I know my room has a dreadful peak at 135 ish and THAT shows up using both mics. But when I reduce it with EQ as REW told me, it just sounds wrong.

Maybe I should have used cardiod pickup instead of omni? i mean, my ears pick up MOSTLY to the front so maybe the graph was too room intensive?

I give up for now. Spent all day on this and made 5% progress.


thanks
ts

 

[truthsurge]

Okay, one more thing.... I noticed in that Cockos EQ that it had a "show phase" checkbox. but there is no indicator that it is a stereo plug or i can eq left/right so when i show phase it adds a line that moves magically based on my boost or cut. ? why??? why me? Does this mean it is a MONO plug? I still hear stereo on the song. I basically want to do this EQ on BOTH speakers or channel material. so even if mono shouldn't it work properly? is it lag related maybe? like it cannot process in time but that makes no sense because I've ran buttloads of tracks in reaper with tons of effects and never had any problems.

So, maybe there is something happening ref phasing that I just don't understand but when i did the up and down of bands maybe it is somehow inducing some phase?

anyway, I'm sure some audio genius knows the problem instantly but then maybe I should use another plug for eq?

grrrrrrrrrrrrr


ts

 

[truthsurge]

SO, I fiddled some more in Reaper and I opened FocusRite Midnight EQ and added a touch of upper mids or low highs like 3K and some high shelf starting at 5 K (say 2 or 3 db each) and boosted 100 about 3 db and to my ears the result sounded a bit clearer with a tad more bass oh I also boosted like 30 a few db. So instead of a song that sounded like there was a thin fleece fabric draped over the speakers, it sounded bright but without noticeable phasing (that I could tell) but then I didn't have 17 bands up 8 to 15 db like I did trying to get a flat response.

So, that brings me to a place where I go Should I be compensating for the highs like this or just find some monitors that sound clearer to begin with? I have the high knob on back of KRK up 1 db and it still was not bright enough. But if I leave the EQ on the master in reaper, yeah it works but then for listening OUTSIDE of reaper, I'm back to the old monitor. Maybe new cards have some built in good eq or maybe there's a good external one you run from soundcard to eq to speakers?

Just trying to get something that will let me get a result that won't suck when someone else listens.

ok.... bedtime. thanks for any tips.

ts

 

[EarlK]

After tons of failures, I hired a team of monkeys who were able to figure out how to loop an output to an input and calibrate my soundcard and then do a measurement of the room and two more up close to my KRK rokit 5s and then the old speaker . But the graphs surprised me. There is a huge scooping of the mids or at least it SEEMS that way for maybe it's not the mids scooped but the lows and highs peaking. ? <snip>

<SNIP>Does anyone ever look at these if I could post them somehow? <SNIP>

Hi,

( Monkeys eh ??? ) well that made me laugh . Glad you got the setup sorted .

Attach the 2 .mdat files to your next post ( the first & second captures ) and I'll take a look at them .

:sn:

 

[Wayne A. Pflughaupt]

Welcome to the forum, truth!

Anyway, I did everything by the tutorial except I didn't have that same mic so I used the AT-10.

That “except” is a make-or-break deal. Without a calibrated mic, you aren’t going to get an accurate graph. You won’t be able to tell if you’re seeing the speaker’s response or the mic’s. I can’t find anything on your mic; everything I see says the AT-10 is a phono cartridge.

I took another reading using my CAD M179 (a good large diaphram mic.
I measured and got a graph that looked similar but better somehow than the AT-10 mic. Didn't look as mid-scooped

So, a measurement with mic “A” looks different from a measurement with mic “B”? You just confirmed the fact that the mic has an influence on the measurements. On top of that, large diaphragm mics aren’t good for response measurements.

This is why you need a calibrated mic suitable for measurements. You really can’t just stick any old mic out there and expect good results. If you want to EQ your speakers, you must first have an accurate graph to work with. Otherwise you’re flying blind.

Regards,
Wayne

 

[truthsurge]

Earl, thanks for the reply and as Wayne says, my graphs are probably off some BUT I know for a fact they are not off completely because both mics gave me similar results so from what I can see, the AT-10 was more sensitive in the 5K to 10K range which is why my graphs of those are more peaked than when I used the CAD M179 (which is a flatter response).

I will try to upload those files and you can at least see them but I know I have to fix the room regardless of what monitors I use. That 135 bump and 100 dip are really in the room. those are not false readings.

So... I will post them here in a bit.

Wayne, you are right! Both mics gave slightly diff readings. But using TWO mics confirmed to me that my graphs were "in the ballpark" and being only a day old in this "room eq" game, I'd call that progress. And making only about $12K per year income, I have to try to work with what I have so buying a new mic just for this IF I could get very close with what I had seemed like a waste of $. I know, if you'd just buy the mic you'd know exactly what your room needs. But now between the TWO mics, I can tell within say 80% of what I need. I need to reorg the contents of my room and prob make some bass traps because the low end is my problem. high end seems a lot better as I have foam on the walls and ceiling and carpet.

So, I have some work to do and unfortunately, I don't pull in 6 figures so I have to spend my $ where I can. That's why I d/l REW.

So, just saying all that so you have a clue where I am coming from.

Thanks for your feedback! I will upload those graphs (about 5 of them using both mics).

Thanks!!!

TS

 

[truthsurge]

Wayne, just a follow up to the "any old mic" statement.

The mics I used are not SM57s, for example. They have a pretty flat response and here is the CAD M179 response curve in omni mode attached. There are some bumps up in the 5k up region but to me, the stuff in the bass area looks flatter than a board. That's where I'm having issues so I am not sure why this mic would be a poor choice compared to the calibrated one if they are both flat response.

But perhaps my questioning like this is perceived as arguing so I just ask because in my brain, things must add up and if a flat response doesn't work for this, why? What is the secret to THAT mic instead of the one I'm using aside from the bumps in the highs which are not a concern for me since my room is dead in the high mids on up? Just asking!

Thanks for everyone's help! I'll get this room better but it will take me some time reading, asking and thinking. and testing.


ts

 

[truthsurge]

Earl,

Here are my files. thanks. I think there are notes in them telling you what mic and if it was room or close to the speaker. I think the CADM179 measurement will be more accurate (although peaked some in the 4K up area due to mic sensitivity in that area).

The 100 dip and 135 peak are horrible, I know. So, maybe you can suggest something but I also have a few desks in there so maybe the bass is amplified and cut based on the sizes of those desk spaces? like everything has a resonant frequency (or more) so maybe if i took that one desk out, it'd help. ?

I have foam on walls and ceiling but no bass traps.


ts

 

[EarlK]

Hi ts,


I got the files ( thanks !) and will comment on them either tomorrow or the next day when I have the time .

Until then, you should create a ( frequency response ) reference point for your speakers ( where the room effect is minimized ( sweep a single speaker ) .
- Close mic one ( from @ 1 ft away using your flattest omni mic / whichever that one is ) .
- Post the .mdat file when it's done, please .

You ought to readup on making a custom calibration file for your flattest omni mic .
- the necessary info is here on the forum within the mic calibration sticky .

:sn:

 

[truthsurge]

Earl,

Thanks! Hey, I did do a close speaker measurement and it is IN one of those files (the one with multiple measurements) but I did NOT do it with my CAD mic so I can't be sure it's the best of the two and my hunch is that it is not as that freq response image I found of my CAD in omni mode is pretty much totally flat up to about 4K which is great because I think almost all my problems are lows to low mids.

So, since that mic is flat up to about 4K, why don't I try a new single-speaker measurement up close at a lower SPL (to eliminate even more room) and then I will have that. But this speaker has woofer and tweeter so 1 ft away pointing at the middle? the woofer? tweeter? and ideally I should probably NOT use omni because that will pick up from all angles. I should use probably Hyper cardioid pattern to get mostly the front. yeah, good idea. heheh

okay, well, another thing I did is to simply by the Ax5 monitors. pricey but I sold some JBL LSR 28Ps in mint shape so it's a wash. Those 28P were just too big and loud for my room.


So, once I get those new monitors I can compare (I know, the room is still not right) but the room won't change MUCH and I can hear which one I like but at $1000 a pair, those A5X already sound good. HEHEHEH


Thanks for all the help. I know I am rambling and it's likely tedious reading this but I just like to pass on what I've done. Maybe a pic of my room as it is now would also help?


anyway, will try to do a measurement close micing the speaker and see. But w/o the mic cal offset, you will see the highs going up past 4k.


Thanks a lot!!!! I feel as if I'm making SOME progress even if tiny. Hope I didn't offend Wayne asking questions. I can be like a kid sometimes asking the dreaded "why" questions. WHY can a large diaphram mic not work well if it has a flat response?



Ts